Method and System for Classifying Document Images

2. The method of claim 1, wherein the numerical coefficients of each of the image segments is calculated as one of a mean value, a standard deviation for at least one of vertical and horizontal directions, and a coordinate of center of mass of the one-dimensional array of pixels.

3. The method of claim 1, wherein each of the sub-codes are generated as a value of standard deviation function over the one-dimensional array of pixels of the image segment.

4. The method of claim 1, wherein each of the sub-codes are generated as a ratio of two standard deviations, calculated for the one-dimensional array of pixels in vertical direction and the one-dimensional array of pixels in horizontal direction.

5. The method of claim 1, wherein the sub-codes are integer numbers that reflect graphical attributes of the image segments of the array or a ratio of attributes between the image segments of the array.

6. The method of claim 1, wherein the image features further include lines, tables, images, and blank spaces.

7. The method of claim 1, further comprising generating similarity rates between at least two document images, each of the document images has a classification code comprised of a combination of sub-codes, wherein the similarity rate is generated by subtracting sub-codes contained in the classification codes of the document images that are to be compared.

8. The method of claim 7, further comprising categorizing the document images based on the similarity rates, wherein the document images with higher similarity rates are placed in a group of similar document layouts.

10. The apparatus of claim 9, wherein the computing device sequentially processes a number of the document images received from the inputting device to generate and store a number of classification codes in the database.

11. The apparatus of claim 9, wherein the numerical coefficients of each of the image segments is calculated as one of a mean value, a standard deviation for at least one of vertical and horizontal directions, and a coordinate of center of mass of the one-dimensional array of pixels.

12. The apparatus of claim 9, wherein the image processor further uses graphical filters as gray-scale thresholds when extracting the image features of said each image segment of the array.

16. The method of claim 15, further comprising categorizing the document images based on the similarity rates, wherein the documents with high similarity rates are grouped in a specific category.

17. The method of claim 15, wherein each of the image segment comprises an array of pixels in two-dimensional directions, the two-dimensional array of pixels is transformed into a one-dimensional array of pixels in either vertical or horizontal direction, and the numerical coefficients of each of the image segments is calculated as one of a mean value, a standard deviation for at least one of vertical and horizontal directions, and a coordinate of center of mass of the one-dimensional array of pixels.

18. The method of claim 15, further comprising applying graphical filters as gray-scale thresholds when processing the image segments of the document images.

19. The method of claim 15, wherein the sub-codes are integer numbers that reflect graphical attributes of the image segments of the array or a ratio of attributes between the image segments of the array.